This year, ~62,000 Americans will be diagnosed with kidney cancer and more than 14,000 individuals will die from this disease. Nine of ten kidney cancers are renal cell carcinoma (RCC). To reduce mortality from RCC, improvements are needed at all stages, from diagnosis to prognosis to therapy. In response to the funding opportunity ?Biological Comparisons in Patient-derived Models of Cancer (U01)?, we will compare four types of patient-derived models of RCC to investigate the relative authenticity of each as a preclinical model. The first will be patient-derived xenografts (PDXs), widely perceived as the most representative models of human pathophysiology. These have been previously established from a range of pathologic and clinical stages of RCC. The PDXs will serve as the ?gold standard? to which to compare three PDX-derived models, including tissue slice cultures (TSCs), primary cell cultures, and xenografts generated from cell cultures. The biological comparison on which we focus is metabolism. Dysregulated metabolism, one of the hallmarks of cancer, is strongly implicated in the development and progression of RCC. Pleiotropic changes include dysregulation of oxygen sensing, energy sensing and nutrient sensing. In particular, high frequency mutations in VHL and FBP1 genes contribute to exhibition of the ?Warburg effect? (an elevation of glycolysis in the presence of oxygen) in clear cell RCC, the major subtype of RCC, leading to increased production and excretion of lactate. Comparing metabolism among the four patient-derived models of RCC will capture the functional consequences of genetic, transcriptomic, environmental and other influences to provide a comprehensive picture of the phenotype of each model system. We will use hyperpolarized (HP) 13C magnetic resonance (MR), a remarkably sensitive molecular imaging technique, to surveil dynamic pathway-specific metabolic and physiologic processes in the patient-derived RCC models, yielding biologically and clinically relevant data. Aim 1 will identify the metabolic signature of each of 8 RCC PDXs by HP MR imaging and steady state metabolomic profiling. The metabolic data will be associated with genotypic, transcriptomic and immunotypic features to establish the phenotype of each PDX. In Aim 2, thin precision-cut tissue slices will be prepared from each of the 8 PDXs and placed in a NMR-compatible, 3D tissue culture bioreactor. The metabolic phenotype of the TSCs will be determined by HP MR and steady state studies and compared to that of the original PDXs, along with genetic, transcriptomic and immunohistologic features. Similar studies will be performed in Aim 3 with primary cell cultures derived from PDXs, and in Aim 4 with xenografts generated by the implantation in mice of PDX-derived cell cultures. In Aim 5, the final test of the four types of models will be a comparison of metabolic responses to the clinically relevant glutaminase inhibitor CB-839, which is currently entering clinical trials in RCC.